This invention relates to a process for detecting nucleotide variations, mutations and polymorphisms by amplifying nucleic acid sequences suspected of containing such mutations or polymorphisms and detecting them with sequence-specific oligonucleotides in a dot blot format.
In recent years, the molecular basis of a number of human genetic diseases has been elucidated by the application of recombinant DNA technology. In particular, the detection of specific polymorphic restriction sites in human genomic DNA associated with genetic disease, such as sickle-cell anemia, has provided clinically valuable information for prenatal diagnosis. In these studies, the presence or absence of a specific site is revealed by restriction fragment length polymorphism (RFLP) analysis, a method in which variation in the size of a specific genomic restriction fragment is detected by Southern blotting and hybridization of the immobilized genomic DNA with a labeled probe. RFLP analysis has proved useful in the direct detection of polymorphic sites that contain the mutation conferring the disease phenotype (e.g., MstII and sickle-cell anemia) as well as in linkage studies where a particular allelic restriction site is linked to a disease locus within a family but not necessarily in the general population. See, for example, Kan and Dozy, Proc. Natl. Acad. Sci. USA, 75, 5631 (1978), and Rubin and Kan, Lancet, 1985-I, 75 (1985). See also Geever et al., Proc. Natl. Acad. Sci. USA, 78, 5081 (1981) and Wilson et al., Proc. Natl. Acad. Sci. USA, 79, 3628 (1982).
In a second method, called "oligomer restriction", a synthetic end-labeled oligonucleotide probe is annealed in solution to the target genomic DNA sequence and a restriction enzyme is added to cleave any hybrids formed. This method, the specificity of which depends on the ability of a base pair mismatch within the restriction site to abolish or inhibit cleavage, is described more fully by Saiki et al., Biotechnology, 3, 1008-1012 (1985). In addition, the sensitivity of this technique may be enhanced by utilizing a polymerase chain reaction procedure wherein the sample is first subjected to treatment with specific primers, polymerase and nucleotides to amplify the signal for subsequent detection. This is described more fully by Saiki et al., Science, 230, 1350-1353 (1985) and in U.S. Pat. No. 4,683,202.
A third method for detecting allelic variations which is independent of restriction site polymorphism utilizes sequence-specific synthetic oligonucleotide probes. See Conner et al., Proc. Natal. Acad. Sci. USA, 80, 78 (1983). This latter technique has been applied to the prenatal diagnosis of .alpha.l-antitrypsin deficiency (Kidd et al., Nature, 304, 230 (1983)) and .beta.-thalassemia (Pirastu et al., N. Eng. J. Med., 309, 284 (1983)). In addition, the technique has been applied to study the polymorphism of HLA-DR.beta. using Southern blotting (Angelini et al., Proc. Natl. Acad. Sci. USA, 83, 4489-4493 (1986)).
The basis for this procedure is that under appropriate hybridization conditions a short oligonucleotide probe of at least 19 bases (19-mer) will anneal only to those sequences to which it is perfectly matched, a single base pair mismatch being sufficiently destabilizing to prevent hybridization. The distinction between the allelic variants is based on the thermal stability of the duplex formed between the genomic DNA and the oligonucleotide (19-mer) probe.
In addition, methods for detecting base pair mismatches in double-stranded RNA and RNA:DNA heteroduplexes have been described using pancreatic ribonuclease (RNase A) to cleave the heteroduplexes. Winter et al., Proc. Natl. Acad. Sci. USA, 82:7575-7579 (1985) and Myers et al., Science, 23(230:1242-1246 (1985). The principal deficiency of this method is its inability to recognize all types of base pair mismatches.
Both the RFLP and oligonucleotide probe stability methods are relatively complex procedures, requiring restriction enzyme digestion, gel-fractionation of the genomic DNA, denaturation of the DNA, immobilization of the DNA either by transfer to a filter membrane or dessication of the gel itself, and hybridization of a labeled probe to the electrophoretically resolved array of immobilized genomic restriction fragments. These steps are necessary, for the oligonucleotide probe stability method, due to the complexity of human genomic DNA. Restriction and electrophoresis are necessary to separate the target sequence ("signal") from the rest of the genome ("noise"), and hybridization in the gel (instead of filter transfer) is necessary to retain as much target sequence as possible. Even then, detection of a signal in a 10 .mu.g sample using a high specific activity kinased probe requires an autoradiographic exposure of four days.
In addition, the approach of Conner et al. requires at least a 19-mer probe for reasons of specificity (a shorter probe would hybridize to more genomic fragments), as well as possibly sensitivity. Shorter probes (e.g., 16-mers), however, would show more sequence-specific discrimination because a single mismatch would be more destabilizing.
There is a need in the art for a simplified method with improved sensitivity and specificity for directly detecting single-base differences in nucleic acids such as genomic DNA.